INDIAN

 INSTITUTE  OF  SCIENCE  

STOCHASTIC HYDROLOGY
Lecture -24
Course Instructor : Prof. P. P. MUJUMDAR
Department of Civil Engg., IISc.
Summary  of  the  previous  lecture  

•  Markov chains
•  Examples on Markov chains

2  
FREQUENCY ANALYSIS
 Frequency Analysis
•  Hydrologic systems are influenced by extreme
events.
–  e.g., severe storms, floods, droughts.
•  The magnitude of an extreme event is inversely
proportional to its frequency of occurrence (i.e.,
more severe events occur less frequently).
•  Frequency analysis is a procedure for estimating
the frequency (or the probability) of occurrence of
extreme events

4  
 Frequency Analysis
•  Objective of frequency analysis of hydrologic data
is to relate the magnitude of extreme events to their
frequency of occurrence using probability
distributions.
•  Hydrologic data to be analyzed is assumed to be
independent and identically distributed and the
hydrologic system is assumed to be stochastic,
space-independent and time independent
•  The data should be properly selected so that the
assumptions of independence and identical
distributions are satisfied.

5  
 Frequency Analysis
•  The assumption of identical distribution or the
homogeneity is achieved by selecting the
observations from same population (i.e., no
changes in the watershed and recording gauges
are made)
•  The assumption of independence is achieved by
selecting the annual maximum of the variable being
analyzed as the successive observations from year
to year will be independent.

6  
 Frequency Analysis
•  The results of flood frequency analysis can be used
for many engineering purposes
–  e.g., flood flow frequency analysis can be used
in the design of dams, bridges, culverts, flood
controlling devices
–  Urban flooding : design storms
–  Drought frequency and magnitude for
agricultural planning

7  
 Frequency Analysis
Return Period
•  An extreme event is defined to have occurred if a
random variable X is greater than (or equal to) a
level xT.
•  The time between the occurrences of X > xT is
called the recurrence interval (τ).
•  The expected value of τ, E(τ), is the average
number of years in which the event X > xT returns.
•  E(τ) is the return period T of the event X > xT.
•  The concept of return period is used to describe the
likelihood of occurrences.
8  
 Frequency Analysis
Example – 1: Consider the annual maximum discharge
Q, (in cumec) of a river for 45 years
Year Q Year Q Year Q Year Q
1950 804 1961 507 1972 1651 1983 1254
1951 1090 1962 1303 1973 716 1984 430
1952 1580 1963 197 1974 286 1985 260
1953 487 1964 583 1975 671 1986 276
1954 719 1965 377 1976 3069 1987 1657
1955 140 1966 348 1977 306 1988 937
1956 1583 1967 804 1978 116 1989 714
1957 1642 1968 328 1979 162 1990 855
1958 1586 1969 245 1980 425 1991 399
1959 218 1970 140 1981 1982 1992 1543
1960 623 1971 49 1982 277 1993 360
1994 348 9  
 Frequency Analysis
Time series of the data:
3500  

3000  
Discharge Q, in Cumec

2500  

2000  

1500  

1000  

500  

0  
1950   1960   1970   1980   1990  
Year

10  
 Frequency Analysis
If xT = 1500cumec, during the period, annual maximum
discharge Q, exceeded xT nine times
Year Q Year Q Year Q Year Q
1950 804 1961 507 1972 1651 1983 1254
1951 1090 1962 1303 1973 716 1984 430
1952 1580 1963 197 1974 286 1985 260
1953 487 1964 583 1975 671 1986 276
1954 719 1965 377 1976 3069 1987 1657
1955 140 1966 348 1977 306 1988 937
1956 1583 1967 804 1978 116 1989 714
1957 1642 1968 328 1979 162 1990 855
1958 1586 1969 245 1980 425 1991 399
1959 218 1970 140 1981 1982 1992 1543
1960 623 1971 49 1982 277 1993 360
1994 348
11  
 Frequency Analysis
The recurrence interval is as shown

Recurrence interval in During the period,

Exceeded year
years the recurrence
1952 interval is ranging
1956 4 from 1 to 14 years;
1957 1 there are 8
1958 1 recurrence intervals
1972 14 covering a total
1976 4 period of 40 years
1981 5 between the first
1987 6 and last
1992 5 occurrences of the
event.
12  
 Frequency Analysis
The return period for 1500 cumec annual maximum
discharge on the river is equal to the average
recurrence interval

40
τ = = 5 years
8
The return period of a given magnitude is defined as
the average recurrence interval between events
equaling or exceeding a specified magnitude

13  
 Frequency Analysis
The probability p = P(X > xT) of occurrence of the
event X > xT in any observation is related with return
period as follows:

For an observation, two outcomes are possible:

success or failure

Success is the probability p i.e., X > xT ; Failure is

probability (1 – p) i.e., X < xT

14  
 Frequency Analysis
Since the observations are independent, the probability
of a recurrence interval of duration τ is the product of
probabilities of τ – 1 failures followed by a success i.e.,
(1 – p)τ-1 p.

1 2 τ-1 τ

(1 – p) (1 – p) …………………………….. p

15  
 Frequency Analysis
The expected value of τ is
∞
τ −1
E (τ ) = ∑τ (1 − p ) p
τ =1

Expanding the equation,

2 3
E (τ ) = p + 2 (1 − p ) p + 3 (1 − p ) p + 4 (1 − p ) p + ...

{ 2 3
= p 1 + 2 (1 − p ) + 3 (1 − p ) + 4 (1 − p ) + .... }
(a)

16  
 Frequency Analysis
The expression with in the brackets has form of power
series expansion

n n ( n − 1) 2 n ( n − 1)( n − 2 ) 3
(1 + x ) = 1 + nx + x + x + ....
2! 3!
With x = –(1 – p) and n = –2.
Then equation (a) becomes
−2
E (τ ) = p (1 − {1 − p})

17  
 Frequency Analysis
p
E (τ ) = 2
(1 − {1 − p})
p
= 2
( p)
1
=
p
E(τ) is the return period T
1
T=
p
18  
 Frequency Analysis
or 1
p=
T
i.e., the probability of occurrence of an event in any
observation is the inverse of its return period
1
P ( X ≥ xT ) =
T

Therefore the T year return period event is X > xT and

it occurs on an average once in T years.

19  
 Frequency Analysis
Example – 2: What is the probability that the annual
maximum discharge in the river will exceed or equal
1500 cumec for the data in Example-1

Sol: Return period of the event [X > 1500] is 5 years

1
P ( X ≥ 1500 ) =
T
1
=
5
= 0.2
20  
 Frequency Analysis
A useful question to be answered is: what is the
probability that a T year return period event will occur
at least once in N years?
• Consider the situation that in N years, the event X >
xT does not occur at all.

• The probability is
(1 – p) x (1 – p) x (1 – p) ……. N times = (1 – p)N

(1 – p) is the probability that in a year the

event X > xT does not occur  

21  
 Frequency Analysis
The complimentary event is that the T year event
occurs at least once in N year period; and hence the
probability is 1 – (1 – p)N

Since p = 1/T,
N
⎛ 1 ⎞
Required probability is 1 − ⎜1 − ⎟
⎝ T ⎠

22  
 Frequency Analysis
Example-3: Obtain the probability that the annual
maximum discharge in the river will equal or exceed
1500 cumec at least once in the next five years
N
⎛ 1 ⎞
Required probability= 1 − ⎜1 − ⎟
⎝ T ⎠
5
⎛ 1 ⎞
= 1 − ⎜1 − ⎟
⎝ 5 ⎠
= 0.672

23  
 Frequency Analysis
Hydrologic data series:
•  Complete duration series: A series containing all
the available data.
•  Partial duration series: A series of data which are
selected so that their magnitude is greater than a
predefined base value.
•  Annual exceedence series: If the base value is
selected so that the number of values is equal to
the number of years.
•  Extreme value series: Series including the largest
or smallest values occurring in each of the equally
long time intervals of the record.
24  
 Frequency Analysis
Hydrologic data series:
•  Annual maximum (or minimum) series: Series with
largest (or smallest) annual values.
•  The return period Te of the event developed from
an annual exceedence series is related with the
corresponding return period T derived from an
annual maximum series by (Chow, 1964)
−1
⎧ ⎛ T ⎞ ⎫
Te = ⎨ln ⎜ ⎟ ⎬
⎩ ⎝ T − 1 ⎠ ⎭
Ref: Chow, V.T., Statistical and probability analysis of hydrologic data sec,8.1 in handbook of
applied hydrology, McGraw Hill, New York, 1964
25  
 Frequency Analysis
•  Annual exceedence series may be difficult to verify
that all the observations are independent
•  ---------------------------------------------
•  Usually annual maximum series is preferred to use.
•  As the return period becomes large, the results
from the two approaches become similar as the
chance that two such events will occur within any
year is very small.

26  
 Frequency Analysis
Extreme Value Distributions:
•  Extreme events:
–  Peak flood discharge in a stream
–  Maximum rainfall intensity
–  Minimum flow
•  The study of extreme hydrologic events involves the
selection of largest (or smallest) observations from
sets of data
–  E.g., the study of peak flows of a stream uses the
largest flow value recorded at gaging station
each year.
27  
 Frequency Analysis
•  Three types of Extreme Value distributions are developed
based on limited assumptions concerning parent
distribution
Ø Extreme value Type-I (EV I) – parent distribution
unbounded in direction of the desired extreme and all
moments of the distribution exist.
Ø  Extreme value Type-II (EV II) – parent distribution
unbounded in direction of the desired extreme and all
moments of the distribution do not exist.
Ø Extreme value Type-III (EV III) – parent distribution
bounded in direction of the desired extreme.

28  
 Frequency Analysis
Extreme Value Type-I (EV I) distribution
•  The cumulative probability distribution function is

⎧ ⎛ x − β ⎞ ⎫ –∞ < x < ∞
F ( x) = exp ⎨− exp ⎜ − ⎟ ⎬
⎩ ⎝ α ⎠ ⎭
•  Parameters are estimated as

β = x − 0.577α
6×s
α=
π
29  
 Frequency Analysis
•  Y = (X – β)/ α → transformation

•  The cumulative probability distribution function is

F ( y) = exp {− exp ( − y )} –∞ < y < ∞

•  Solving for y,
⎧⎪ ⎛ 1 ⎞ ⎫⎪
y = − ln ⎨− ln ⎜
⎜ F ( y ) ⎟⎟ ⎬
⎪⎩ ⎝ ⎠ ⎪⎭

30  
 Frequency Analysis
1 1
P ( X ≥ xT ) = ; 1 − P ( X < xT ) =
T T
1 1
1 − F ( xT ) = ; F ( xT ) = 1 −
T T
T −1
=
T
Therefore

⎧ ⎛ T ⎞ ⎫
y = − ln ⎨ln ⎜ ⎟ ⎬
⎩ ⎝ T − 1 ⎠ ⎭

31  
 Example – 1
Consider the annual maximum discharge of a river for
45years in the previous example,
1.  develop a model for annual maximum discharge
frequency analysis using Extreme Value Type-I
distribution and
2.  calculate the 20 year and 100 year return period
maximum annual discharge values

32  
 Example – 1 (Contd.)
Data is as follows:

Year Data Year Data Year Data Year Data

1950 804 1961 507 1972 1651 1983 1254
1951 1090 1962 1303 1973 716 1984 430
1952 1580 1963 197 1974 286 1985 260
1953 487 1964 583 1975 671 1986 276
1954 719 1965 377 1976 3069 1987 1657
1955 140 1966 348 1977 306 1988 937
1956 1583 1967 804 1978 116 1989 714
1957 1642 1968 328 1979 162 1990 855
1958 1586 1969 245 1980 425 1991 399
1959 218 1970 140 1981 1982 1992 1543
1960 623 1971 49 1982 277 1993 360
1994 348
33  
 Example – 1 (Contd.)
Mean of the data, x = 756.6 cumec
Standard deviation, s = 639.5 cumec
6×s
α=
π
6 × 639.5
=
π
= 498.6
β = x − 0.577α
= 756.6 − 0.577 × 498.6
= 468.8
34  
 Example – 1 (Contd.)
The probability model is
⎧ ⎛ x − 468.8 ⎞ ⎫
F ( x) = exp ⎨− exp ⎜ − ⎟ ⎬ –∞ < x < ∞
⎩ ⎝ 498.6 ⎠ ⎭

To determine the xT value for a particular return period,

the reduced variate y is initially calculated for that
particular return period using
⎧ ⎛ T ⎞ ⎫
y = − ln ⎨ln ⎜ ⎟ ⎬
⎩ ⎝ T − 1 ⎠ ⎭

35  
 Example – 1 (Contd.)
For T = 20 years,
⎧ ⎛ 20 ⎞ ⎫
y20 = − ln ⎨ln ⎜ ⎟ ⎬
⎩ ⎝ 20 − 1 ⎠ ⎭
= 2.97

x20 = β + α y20
= 468.8 + 498.6 * 2.97
= 1950 cumec

36  
 Example – 1 (Contd.)
Similarly for T = 100 years,
⎧ ⎛ 100 ⎞ ⎫
y20 = − ln ⎨ln ⎜ ⎟ ⎬
⎩ ⎝ 100 − 1 ⎠ ⎭
= 4.6

x20 = β + α y20
= 468.8 + 498.6 * 4.6
= 2762 cumec

37  
 Frequency Analysis
Frequency analysis using frequency factors:
•  Calculating the magnitudes of extreme events by
the above prescribed method requires that the
cumulative probability distribution function to be
invertible.
•  Some probability distribution functions like normal,
Pearson type-III distributions are not readily
invertible.
•  The alternative method of calculating the
magnitudes of extreme events is using frequency
factors

38  
 Frequency Analysis
•  The magnitude of an event xT is represented as the
mean µ plus the deviation ΔxT of the variate from
the mean
xT = µ + ΔxT

•  The deviation is taken as equal to the product of

standard deviation σ and a frequency factor KT
ΔxT = KT σ
x T = µ + KT σ
or
xT = x + KT s
39  
 Frequency Analysis
•  The deviation and the frequency factor are
functions of the return period and the type of
probability distribution to be used in the analysis
•  If in the event, the variable analyzed is y = logx,
then the same method is used to the statistics for
the logarithms of the data as
yT = y + KT s y
•  For a given distribution, a K – T relationship can be
determined
•  The relationship can be expressed in terms of
mathematical terms or a table.

40  
 Frequency Analysis
The procedure for frequency analysis is as follows

•  Obtain the statistical parameters required for the

probability distribution from the given data
•  For a given return period T, the frequency factor is
determined from the K – T relationship proposed for
the distribution
•  The magnitude of xT is then computed by

xT = x + KT s

41  
 Frequency Analysis
Frequency factor for Normal Distribution:

•  The frequency factor is expressed as (same as the

standard normal deviate z)

2.515517 + 0.802853w + 0.01032w2

KT = w −
1 + 1.432788w + 0.189269w2 + 0.001308w3
where 1
2
⎡ ⎛ 1 ⎞ ⎤
w = ⎢ln ⎜ 2 ⎟ ⎥ 0 < p < 0.5
⎣ ⎝ p ⎠ ⎦
42  
 Frequency Analysis
•  When p > 0.5, 1 – p is substituted and the KT value
computed is given an negative sign
•  The error in this formula is less than 0.00045
(Abramowitz and Stegun, 1965)
•  The frequency factor KT for the normal distribution is
equal to standard normal deviate z.
•  The same procedure is applied for the lognormal
distribution expect that the logarithm of data is used
to calculate mean and standard deviation.

43  
 Example – 2
Consider the annual maximum discharge of a river for
45years in the previous example,

The mean of the data, x = 756.6 cumec

Standard deviation, s = 639.5 cumec

Calculate the frequency factor and obtain the

maximum annual discharge value corresponding to 20
year return period using Normal distribution

44  
 Example – 2 (Contd.)
T = 20
p = 1/20 = 0.05
1
2
⎡ ⎛ 1 ⎞ ⎤
w = ⎢ln ⎜ 2 ⎟ ⎥
⎣ ⎝ p ⎠ ⎦
1
2
⎡ ⎛ 1 ⎞ ⎤
= ⎢ln ⎜ 2 ⎟ ⎥
⎣ ⎝ 0.05 ⎠ ⎦
= 2.45

45  
 Example – 2 (Contd.)
2.515517 + 0.802853w + 0.01032 w2
KT = w −
1 + 1.432788w + 0.189269 w2 + 0.001308 w3
2.515517 + 0.802853 × 2.45 + 0.01032 × 2.45 2
= 2.45 −
1 + 1.432788 × 2.45 + 0.189269 × 2.452 + 0.001308 × 2.453
= 1.648

xT = x + KT s
= 756.6 + 1.648 × 639.5
= 1810.5 cumec

46  
 Frequency Analysis
Frequency factor for Extreme Value Type I(EV I)
Distribution:

6 ⎧ ⎡ ⎛ T ⎞ ⎤ ⎫
KT = − ⎨0.5772 + ln ⎢ln ⎜ ⎟ ⎥ ⎬
π ⎩ ⎣ ⎝ T − 1 ⎠ ⎦ ⎭
•  To express T in terms of KT,
1
T=
⎧⎪ ⎡ ⎛ π KT ⎞ ⎤ ⎫⎪
1 − exp ⎨− exp ⎢ − ⎜ 0.5772 + ⎟ ⎥ ⎬
⎪⎩ ⎣ ⎝ 6 ⎠ ⎦ ⎪⎭

47  
 Frequency Analysis
xT − µ
When xT = µ in the equation KT = ; KT = 0
σ
Substituting KT = 0,
1
T=
⎧⎪ ⎡ ⎛ π × 0 ⎞ ⎤ ⎫⎪
1 − exp ⎨− exp ⎢ − ⎜ 0.5772 + ⎟ ⎥ ⎬
⎪⎩ ⎣ ⎝ 6 ⎠ ⎦ ⎪⎭
= 2.33 years
i.e., the return period of mean of a EV I is 2.33 years

48  
 Example – 3
Consider the annual maximum discharge of a river for
45years in the previous example,

The mean of the data, x = 756.6 cumec

Standard deviation, s = 639.5 cumec

Calculate the frequency factor and obtain the

maximum annual discharge value corresponding to 20
year return period using Extreme Value Type I(EV I)
distribution

49  
 Example – 2 (Contd.)
T = 20 years
6 ⎧ ⎡ ⎛ T ⎞ ⎤ ⎫
KT = − ⎨0.5772 + ln ⎢ln ⎜ ⎟ ⎥ ⎬
π ⎩ ⎣ ⎝ T − 1 ⎠ ⎦ ⎭
6 ⎧ ⎡ ⎛ 20 ⎞ ⎤ ⎫
=− ⎨0.5772 + ln ⎢ln ⎜ ⎟ ⎥ ⎬
π ⎩ ⎣ ⎝ 20 − 1 ⎠ ⎦ ⎭
= 1.866
xT = x + KT s
= 756.6 + 1.866 × 639.5
= 1949.9 cumec
50  
 Frequency Analysis
Frequency factor for Log Pearson Type III Distribution:

•  The data is converted to the logarithmic series by

y = logx.
•  Logarithms to base 10 is usually used
•  The mean y , standard deviation sy, and the
coefficient of skewness Cs are calculated for the
converted logarithmic data
•  The frequency factor depend on the return period
and coefficient of skewness

51  
 Frequency Analysis
•  When Cs = 0, the frequency factor is equal to the
standard normal deviate z and is calculated as in
case of Normal deviation.
•  When Cs ≠ 0, KT is calculated by (Kite, 1977)
1 3
( ) ( ) ( )
KT = z + z − 1 k + z − 6 z k 2 − z 2 − 1 k 3
2

3
4 1 5
+ zk + k
3
Where k = Cs / 6

52  
 Example – 3
Consider the annual maximum discharge of a river for
45years in the previous example,

Calculate the frequency factor andx obtain the

maximum annual discharge value corresponding to 20
year return period using Log Person Type III
distribution

The logarithmic data series is first obtained

53  
 Example – 2 (Contd.)
The mean of the data, y = 2.725 cumec
Standard deviation, s = 0.388 cumec
Coefficient of skewness Cs = -0.2664
T = 20 years
1
2
⎡ ⎛ 1 ⎞ ⎤
w = ⎢ln ⎜ 2 ⎟ ⎥
⎣ ⎝ p ⎠ ⎦
1
2
⎡ ⎛ 1 ⎞ ⎤
= ⎢ln ⎜ 2 ⎟ ⎥
⎣ ⎝ 0.05 ⎠ ⎦
= 2.45
54  
 Example – 2 (Contd.)
2.515517 + 0.802853w + 0.01032 w2
z = w−
1 + 1.432788w + 0.189269 w2 + 0.001308 w3
2.515517 + 0.802853 × 2.45 + 0.01032 × 2.45 2
= 2.45 −
1 + 1.432788 × 2.45 + 0.189269 × 2.452 + 0.001308 × 2.453
= 1.648

k = Cs / 6
= -0.2664/6
= -0.0377

55  
 Example – 2 (Contd.)
1 2
( )
KT = 1.648 + 1.6482 − 1 ( −0.0377 ) +
3
( )
1.6483 − 6 × 1.648 ( −0.0377 )

3 4 1 5
( 2
)
− 1.648 − 1 ( −0.0377 ) + 1.648 × ( −0.0377 ) + ( −0.0377 )
3

= 1.581

xT = x + KT s
= 756.6 + 1.581× 639.5
= 1767.6 cumec

56  
PROBABILITY PLOTTING
 Probability Plotting
•  To check whether probability distribution fits a set of
data or not the data is plotted on specially designed
probability paper.
•  The cumulative probability of a distribution is
represented graphically on probability paper
designed for the distribution.
•  The plot is prepared with exceedence probability or
the return period T on abscissa and the
magnitude of the event on ordinate.
•  The scales of abscissa and ordinate are so
designed that the data to be fitted are expected to
appear close to straight line
58  
 Probability Plotting
•  The purpose of using the probability paper is to
linearize the probability relationship
•  The plot can be used for interpolation, extrapolation
and comparison purposes.
•  The plot can also be used for estimating
magnitudes with other return periods.
•  If the plot is used for extrapolation, the effect of
various errors is often magnified.

59  
 Plotting Position
•  Plotting position is a simple empirical technique
•  Relation between the magnitude of an event verses
its probability of exceedence.
•  Plotting position refers to the probability value
assigned to each of the data to be plotted
•  Several empirical methods to determine the plotting
positions.
•  Arrange the given series of data in descending
order
•  Assign a order number to each of the data (termed
as rank of the data)
60  
 Plotting Position
•  First entry as 1, second as 2 etc.
•  Let n is the total no. of values to be plotted and
m is the rank of a value, the exceedence
probability (p) of the mth largest value is obtained
by various formulae.
•  The return period (T) of the event is calculated by
T = 1/p
•  Compute T for all the events
•  Plot T verses the magnitude of event on semi log or
log log paper

61  
 Plotting Position
Formulae for exceedence probability:

California Method:

m
P ( X ≥ xm ) =
n
Limitations
–  Produces a probability of 100% for m = n

62  
 Plotting Position
Modification to California Method:
m −1
P ( X ≥ xm ) =
n

Limitations
–  Formula does not produce 100% probability
–  If m = 1, probability is zero

63  
 Plotting Position
Hazen s formula:

m − 0.5
P ( X ≥ xm ) =
n

Chegodayev s formula:
m − 0.3
P ( X ≥ xm ) =
n + 0.4

Widely used in U.S.S.R and Eastern European

countries
64  
 Plotting Position
Weibull s formula:
–  Most commonly used method
–  If n values are distributed uniformly between 0
and 100 percent probability, then there must be n+1
intervals, n–1 between the data points and 2 at the
ends.
m
P ( X ≥ xm ) =
n +1

–  Indicates a return period T one year longer than the

period of record for the largest value
65  
 Plotting Position
Most plotting position formulae are represented by:
m−b
P ( X ≥ xm ) =
n + 1 − 2b
Where b is a parameter

–  E.g., b = 0.5 for Hazen s formula, b = 0.5 for

Chegodayev s formula, b = 0 for Weibull s formula
–  b = 3/8 0.5 for Blom s formula
–  b = 1/3 0.5 for Tukey s formula
–  b = 0.44 0.5 for Gringorten s formula

66  
 Plotting Position
–  Cunnane (1978) studied the various available
plotting position methods based on unbiasedness
and minimum variance criteria.
–  If large number of equally sized samples are
plotted, the average of the plotted points foe each
value of m lie on the theoretical distribution line.
–  Minimum variance plotting minimizes the variance
of the plotted points about the theoretical line.
–  Cunnane concluded that the Weibull s formula is
biased and plots the largest values of a sample at
too small a return period.

67  
 Plotting Position
–  For normally distributed data, the best formula is
Blom s plotting position formula (b = 3/8).
–  For Extreme Value Type I distribution, the
Gringorten formula (b = 0.44) is the best.

68  
 Example – 3
Consider the annual maximum discharge of a river for
45years, plot the data using Weibull s formula
Year Data Year Data Year Data Year Data
1950 804 1961 507 1972 1651 1983 1254
1951 1090 1962 1303 1973 716 1984 430
1952 1580 1963 197 1974 286 1985 260
1953 487 1964 583 1975 671 1986 276
1954 719 1965 377 1976 3069 1987 1657
1955 140 1966 348 1977 306 1988 937
1956 1583 1967 804 1978 116 1989 714
1957 1642 1968 328 1979 162 1990 855
1958 1586 1969 245 1980 425 1991 399
1959 218 1970 140 1981 1982 1992 1543
1960 623 1971 49 1982 277 1993 360
1994 348
69  
 Example – 3 (Contd.)
•  The data is arranged in descending order
•  Rank is assigned to the arranged data
•  The probability is obtained using
m
P ( X ≥ xm ) =
n +1

•  Return period is calculated

•  The maximum annual discharge verses the return
period is plotted

70  
 Example – 3 (Contd.)
Annual Arranged
Year Rank (m) P(X > xm) T
Max. Q data
1950 804 3069 1 0.021739 46
1951 1090 1982 2 0.043478 23
1952 1580 1657 3 0.065217 15.33333
1953 487 1651 4 0.086957 11.5
1954 719 1642 5 0.108696 9.2
1955 140 1586 6 0.130435 7.666667
1956 1583 1583 7 0.152174 6.571429
1957 1642 1580 8 0.173913 5.75
1958 1586 1543 9 0.195652 5.111111
1959 218 1303 10 0.217391 4.6
1960 623 1254 11 0.23913 4.181818
71  
 Example – 3 (Contd.)

1500  
Annual  Maximum  discharge  (Q)  

150  

15  
1   10   100  
Return  period  (T)  

72